Abstract:
Procedure:
It has become increasingly more important
to be able to correctly forecast large solar
flares because these events can destroy or
interrupt important technology and harm
astronauts. We looked at the relationships
between different measurements and
classifications of sunspots for patterns in
the type of sunspot that produced a large
flare. We selected active regions with
complete data sets containing each of the
measurements we wanted to compare; if an
active region was missing a measurement,
we removed it from the analysis. We also
measured the distance between each active
region and the nearest other active region
on the disk. Once we obtained this
information, we generated plots and
histograms to look at the tendencies and
frequencies of X-class, M-class, and Cclass flares as compared to Non-flaring
active regions. The trends were weak, but
we found a correlation between various
measures of active region complexity and
flaring tendencies. Comparing this to the
NHGV value (a parameter that uses
subsurface motions to predict flare
occurrence) we found that NHGV values
tend to be higher for more complex and
more compact active regions, even when
the active regions produce a similar size
flare.
Within a previously writing IDL program
that accounted for all parameters other than
distance, we wrote a subroutine to extract
the location of an active region and convert
it into usable coordinates. We needed to
convert the longitudinal coordinates into
co-latitudes in order to use spherical
geometry to calculate the distance between
the two active regions. Then, we wrote
another subroutine to measure the distance
between the target active region and the
next nearest active region using the
converted coordinates. After incorporating
the two subroutines, we were able to build
plots to look for correlation patterns.
Introduction:
In this investigation, we look more closely at the properties of active regions and compare flaring
regions with similar non-flaring regions. The properties of the active regions that we investigated
were compactness classification, penumbra classification, Zurich classification, Normalized
Helicity Gradient Variance (NHGV), number of spots in the sunspot group, longitudinal extent,
area of the sunspot group, and distance between the target active region and the next closest active
region. In this investigation, we added the distance parameter in attempt to improve the
forecasting of flare events. Especially with large scale flares, knowing which regions will produce
flares helps us prepare for severe geomagnetic storms, solar radiation storms, and radio blackouts
that could last for days damaging technology and presenting increased health risks for humans.
k, no flare, 1.04
h, no flare, 1.08
h, no flare, 1.08
k, x-class, 1.17
h, x-class, 1.15
h, x-class, 1.15
c, no flare, 0.98
i, no flare, 1.02
d, no flare, 1
e, no flare, 1.03
c, x-class, 1.24
i, x-class, 1.13
d, x-class, 1.13
e, x-class, 1.2
Conclusion:
In conclusion, the more compact a sunspot group is the higher the probability of producing an xclass flare, sunspots with asymmetric penumbra are more likely to flare in the x-class then
sunspots with symmetric or no penumbra. Elongated bipolar sunspot groups with penumbra at
both ends are more likely to flare in the x-class than single spots, those without penumbra, and
uni-polar sunspots. The distance parameter only slightly improved the accuracy but did not
improve the false alarm rate. Studying the relationships between these parameters is still a work
in progress.
http://ve4xm.caltech.edu/Bellan_plasma_page/laborato.htm 3 Aug. 2010
http://www.petermeadows.com/html/area.html 3 Aug. 2010
http://solar‐center.stanford.edu/solar‐images/magnetograms.html 3 Aug. 2010
http://spaceweather.com/glossary/filaments.html 3 Aug. 2010
http://solarmonitor.org/ 3 Aug. 2010
http://spaceweather.com/glossary/flareclasses.html 2 Aug. 2010
http://www.swpc.noaa.gov/info/glossary.html 2 Aug. 2010
http://www.swpc.noaa.gov/info/glossary.html 2 Aug. 2010
http://www.nso.edu/staff/apevtsov/www/gallery.html 2 Aug. 2010
http://solar.physics.montana.edu/magara/Research/Research_sig.html 2 Aug. 2010
http://euromet.meteo.fr/resources/ukmeteocal/verification/www/english/msg/ver_categ_f
orec_ex/uos3/uos3_ko1.htm 30 July 2010
http://www.cawcr.gov.au/projects/verification/ 30 July 2010
Reinard, A., J. Henthorn, R. Komm, and F. Hill. “Evidence that Temporal Changes in Solar Subsurface Helicity Precede Active Region Flaring.” The Astrophysical Journal Letters 20 Feb. 2010: L121‐L125. Print.